1896. NOTES AND COMMENTS. 355 



may not hope for a reliable estimate of the length of the pre-Cambrian 

 period from zoology any more than from physics. We may at least 

 be grateful, when comparing Professor Poulton's cautious generalities 

 with the statement of some physicists, that no attempt to hamper 

 progress by impossible precision is likely to be made. 



Palaeontology and Geographical Distribution. 



A FURTHER example of the greater importance now attached to 

 palaeontology is given by Mr. Lydekker's volume on the " Geographical 

 Distribution of Mammals," on the issue of which we congratulate the 

 editor of the Cambridge Geographical Textbooks. Mr. Lydekker's 

 work, reviewed on p. 392, is unique among manuals on distribution in 

 the attention paid to fossil forms. The works of Schmarda (1853), 

 Sclater (1858), Huxley (1868), and Wallace (i876)established zoological 

 provinces on the evidence of living animals alone. Neglect of 

 palaeontology was necessary when these authors wrote, since the 

 available evidence was so imperfect. The methods they were there- 

 fore forced to employ were as little likely to yield final results, as an 

 attempt to deduce the original distribution of human races from that 

 of the present day without any reference to the historical records of 

 changes and migrations. It is not surprising that Mr. Lydekker's 

 broader range of view renders necessary alterations in the classifica- 

 tion of the zoological regions. Mr. Lydekker finds the time-honoured 

 six regions of Mr. Sclater inadequate, and prefers ten regions divided 

 among three realms, which are those suggested by Dr. Blanford in 

 1890. The ten regions are prepared mainly on the evidence of 

 mammals, except in the case of the Hawaiian, which in the absence of 

 a mammalian fauna depends on birds. The striking differences 

 between arrangements based on different classes of animals them- 

 selves illustrate the necessity for the consideration of palaeontology ; 

 for the differences are due, at least in part, to the varying distribution 

 of land and water in past ages. That Mr. Lydekker's divisions do 

 not suit all classes of animals detracts from their convenience but not 

 from their truth. It may be objected that palaeontological evidence 

 cannot be relied on owing to the imperfection of the geological record. 

 But after the discoveries among the mammalian faunas of Patagonia 

 and the Argentine Republic, of the Western States of America, of India, 

 of Australia, of the Mediterranean islands, not to mention the many 

 localities on the mainland of Europe, this objection may be regarded 

 as equal in value to the cry for missing links. Mr. Lydekker's book 

 shows that systems founded on neontological evidence only have had 

 their day. 



International Congresses. 



It has been decided by the British Association that all fellows 

 and members of the American Association, meeting next year at 



